An investigation of factors predictive of independence in transfers and ambulation after hip fracture

Objectives: To compare the predictive value of measurements of mobility on the second day postsurgery with previously established outcome predictors after hip fracture and to establish a statistical model for the prediction of independence in transfers and ambulation. Design: Prospective, validation cohort study. Setting: Primary care center. Patients: Two samples of 50 patients admitted with primary diagnosis of hip fracture. Interventions: Not applicable. Main Outcome Measures: Independence in transfers and ambulation within 2 weeks of surgery. Predictor variables considered were age, mental state, prefracture mobility, and 4 measurements of transfers and ambulation on the second day postsurgery. Results: In bivariate logistic regression analysis, all variables were significant predictors. In multiple logistic regression analysis, only distance walked and assistance required in transferring supine to sitting on day 2 postsurgery were significant. The multiple logistic regression model produced from the analysis had an outcome classification accuracy of 88% when tested on an independent sample. Conclusions: Measurements of mobility on day 2 postsurgery are significant and reliable predictors of independence in transfers and ambulation. Further consideration of the variables age, mental state, and prefracture mobility do not appear to improve the accuracy of the prediction.